Project Publication on the Web — Addendum II

Andrea Vianello

(See email contacts page for the author's email address.)

The Problem with the Monolingual Websites

As a follow-up on the series of articles on project publication on the Web, I shall discuss here briefly the problem of foreign languages in web publications. Scholars generally publish in their own native language, be it English, French, German, etc. For scholars in some smaller countries, a more widely spoken language might be chosen, but the assumption has been that readers will deal with foreign-language material when the need arises. Indeed, facility with foreign languages is often a requirement of a higher degree. As the web has grown, however, English has become the lingua franca, and publications on the web have become more and more likely to be in English. Even websites produced in non-English-speaking countries often choose English as the base language on the assumption that it is the most widely read and understood modern language. While that assumption may be true, it ignores masses of people who do not speak English or do not read English or cannot read English well enough to understand complex material on the web.

In archaeology the problem is more pressing. As identified already in the original series of the project on Web publishing, it is important to connect with local people, those living near the sites, who for the most part do not speak English and are often ignored in the publication output.

Websites more than any other publication type are international in character. They are immediately accessible to anyone with access to the Internet, without limitations due to academic background, nationality, financial status, or language. Here the language barrier can be a more serious issue by preventing valuable exchanges with foreign-speakers. In addition, a website about a project may, by not engaging local communities, effectively discourage the long-term conservation of archaeological sites. If the local people do not understand the value of the monuments surrounding them, they are deprived of their past and see little value in the long-term conservation of sites or future field research.

In my opinion, we need to change the idea that the English language facilitates communication in all cases, recognize that it can be a potential problem, and take action to include other languages where needed.

The Web and its Opportunities

Search engines such as Google and Bing have initiated subtle ways to break language barriers. They make use of machine translations and lately even suggest keywords in foreign languages to expand searches. By doing so, these search engines enable access to contents in any of the other supported languages. Even more important, it is now possible to display web pages in the language of the reader's choice, regardless of the language in which the page was written. In short, a need that is most pressing for academics and anyone else interested in cultural matters is finding solutions outside academia. If being inclusive is generally useful to all, it should be more so to academics.

Machine translations of websites are increasingly accurate and easily accessible (most likely a single click in the browser will suffice). The accuracy however is based on common words and phrases, and it will dip with academic contents, no matter how they are explained. I recommend therefore writing basic pages in clear, concise, and uncomplicated prose. The resulting pages will serve as introduction, will be aimed at non-specialists, and may be tested with some of the available translation tools. If such texts fail the testing (i.e. translated texts cannot be understood by a foreign speaker or in the case of double translations, to foreign language and back to English, the writer cannot recognize the text), then shorter, simpler sentences and a little additional clarity will be necessary. This is actually a good, yet imperfect, test to determine one’s clarity in writing.

I would recommend going so far as placing a link on some pages to a machine translator that has been tested on those pages; it will limit damage from poor translations for the general introductory texts. (There should be a statement indicating that such translations may be riddled with errors and should be used with care.)

I do not advocate here the need for researchers to write multiple languages and produce multilingual output by themselves. They do need to take some responsibility for making their published output accessible to people speaking languages other than English. Writing core texts in English is still the best solution for providing broad accessibility. However, it is important to be aware that automated translations will be used, whether one likes it or not, and that clear, concise, uncomplicated prose is good for all, readers of English and readers of machine translations from that English.

Although few websites will have the resources to include professional translations, the planning of a website should include discussion of translations into multiple languages, making the cases for and against, and deciding firmly on a policy. Preference should be given to English, if the core text is written in another language, to the language spoken locally (in case of field or regional projects), and to languages with an existing substantial literature for the topic. The most widely spoken languages in the world or those of the collaborators in the project may provide further suggestions for additional languages.

In my view, each major project should be accessible at least in three languages, which seems to me a reasonable proposition. The Rosetta Stone is inscribed in three scripts after all, and all archaeologists will be familiar with the benefits of that choice. Three is the lowest number with a high enough probability that at least one language will be understood (given the criteria above of English plus local language and another major language).

For smaller projects, machine translations should suffice, unless you can write in multiple languages and have the time to do so or have a budget for translation or know someone (colleague, spouse, etc.) who can translate for you.

For larger projects, professional translators should be used. Because of the costs involved, I think the translations should be limited to major texts and should not be updated. Whatever a project director decides, the policy should be stated clearly on the home page of the site. In addition, project personnel must plan ahead, and I think they should use tools such as CMS packages that separate texts from graphics and are multi-language friendly. Those packages make it possible to keep texts in various languages separate from the graphics, layout, and data and then appear integrated. (Alternatively, it is possible to duplicate the textual part of a website in various languages, reduce graphics and layout to a minimum, and publish complete translations not to be altered.)

The main language of any website, the one that will benefit from updates and to be used in citations and as reference, should be clearly stated as well. That will avoid confusion. Readers will be pleased to find a website in a language they can understand and will be forgiving of shortcomings if they know that they are reading a translation to help them, rather than a full and proper publication. Concerns about costs and management of translations can also be addressed more easily if translated pages are not kept rigorously up-to-date. (Costs of translations will depend on the length of text to be translated, but the translations should need no further management unless very serious omissions or errors require revision). I appreciate that translations in multiple languages will add complexity to a website, and that it is simply impossible to manage a website in multiple languages without staff involved in the production of the website actively speaking all of them. I only suggest that researchers decide which languages to use and how to manage the process.

Conclusions

I have presented here some reasons for producing multi-lingual academic websites. The use of multiple languages provides advantages in terms of inclusiveness and accessibility of large groups of people. The use of English as sole language, despite being the most widely spoken language in the world and preferred language for international communication, facilitates targeting the largest group of readers, but can also have the effect of excluding an even larger number of people. English is only spoken by a minority of the seven billion or so speakers in the world. It is naïve to think that everybody can speak English, and the scientific mandate of diffusing the results of research imposes on all scholars the need to communicate the results of research as broadly as possible. Either facilitating translations (by writing carefully) or providing them on a website seems desirable as a result.

Producing academic websites poses many challenges, and, so far, few rewards. Introducing translations adds to the complexity, but it is one of the few things that can produce immediate and tangible benefits. There are good ways to produce translations, which will expand the outreach of a project significantly. There are also bad ways (namely unchecked machine translations) to produce translations, which are already here and will be used if nothing else is available. Websites are already the first point of contact with the public and students, and regardless of what scholars think, they are bound to become the first point of contact for researchers too as more resources become available online. Therefore, scholars truly need to present their research at its best and to the largest audience possible. Multi-lingual websites are ready for the task, and coming to your rescue.

Postscript – an Idea for the Future

Publishing on the Web means making contents more accessible. I believe that eventually academic contents will be grouped and made accessible through specialized websites (Google Scholar and Scirus are leading the way in that direction). In such a long-term view, the time will come when we must loosen up access to contents, and use Creative Commons licenses to make the re-use of texts easier (with due limitations for commercial use and requirements for appropriate citations). Opening up websites also makes possible the involvement of a community of readers for many projects. Some projects may even be aimed squarely at colleagues so that they may contribute ideas or knowledge about certain topics (e.g. suggest bibliography, peer review, etc.). Other projects, however, may promote the translation of websites in multiple languages by the community, at little cost to the project directors. Archaeologists (and other scholars) have yet to tap into this potential resource.

-- Andrea Vianello